Electronic nose identifies wood species and can help to combat illegal logging

Devices developed by Brazilian researchers can also be used for early detection of fungal contamination of oranges and to distinguish between different types of plastic.

By Elton Alisson

Agência FAPESP – Researchers at the University of São Paulo’s Chemistry Institute (IQ-USP) in Brazil have built “electronic noses” capable of identifying and classifying (by odor) different types of wood and plastic as well as detecting early-stage deterioration of oranges due to fungal contamination.

Some of the devices were developed through the project “New conjugated polymers for solar cells and electronic noses,” conducted with support from FAPESP.

“The technology is very simple and inexpensive and has several applications,” said Jonas Gruber, a professor at IQ-USP and principal investigator for the project.

Each “e-nose” consists of an array of gas sensors that change the electrical conductivity of some of the materials that they are made of (including conductive polymer, a type of plastic) as they interact with vapors from volatile substances such as amines, alcohols, ketones and aromatic compounds.

Variations in the array’s conductance generate a specific electrical signal that is converted into a digital signal. This signal is read by a computer program, which, in a matter of seconds, identifies the type of volatile substance in contact with the device.

“The e-nose responds differently according to the nature of the gas that comes into contact with the polymeric materials in the sensors,” Gruber told Agência FAPESP. One feat in particular permitted the development of these e-noses. Gruber and his group at IQ-USP synthesized and characterized new conductive polymers derived from two specific classes of polymer: poly-p-phenylenevinylene (PPV) and poly-p-xylylene (PPX). They then used these to build sensors.

“We were the first to use PPV in gas sensors,” Gruber said. “The advantages are low production cost, low power consumption, and the ease with which the characteristics of the devices can be changed by introducing structural modifications into the polymer chains.”

The sensor construction technique used by the researchers consists of depositing a conductive polymer film with a thickness of a few hundred nanometers (billionths of a meter) on a board the size of a cell phone chip with two interdigitated metal electrodes, so that the film connects the two materials. The interdigitated electrodes are deposited in an interlocking configuration, with minute gaps between them.

When the sensor is exposed to vapor from a volatile substance, the film’s electrical resistance changes. “Each sensor costs one Brazilian Real, or less than three US Dollars, and we use between four and seven sensors per e-nose on average,” Gruber said.

Wood identification

One of the devices was developed to identify and classify different species of wood. The researchers expect it to be used by environmental police to combat illegal logging of endangered tree species in Brazil’s tropical forests.

Protected species such as mahogany (Swietenia macrophylla) are often hard to distinguish from species that can be legally logged and marketed, such as Spanish cedar (Cedrela odorata).

Both belong to the same family (Meliaceae) and are so similar that mahogany is often marketed as Spanish cedar.

“You can distinguish between mahogany and Spanish cedar when you look at them in the forest, but once logged, they can be distinguished only by histological analysis performed in the lab by a botanist,” Gruber said.

E-noses facilitate the identification of these and other species, such as Brazilian walnut (Ocotea porosa) and Ocotea catharinensis (local common name: canela-preta). All they need is a small shaving from the trunk, which releases volatile compounds. The sensor array identifies the species in under a minute.

“Because Spanish cedar and mahogany are different species and belong to different genera, e-noses can distinguish them with 100% accuracy,” Gruber said. “In the case of walnut and canela-preta, it’s a little harder because they belong to the same genus. Even so, identification by e-nose is correct 95% of the time.”

Aged cachaça

The e-nose for wood identification aroused the interest of researchers at the Aguardente Chemistry Development Laboratory (LDQA), part of the University of São Paulo’s São Carlos Chemistry Institute. They wanted a way to distinguish cachaça aged in oak casks from cachaça aged in barrels composed of other wood species that are considered inferior. Cachaça is made from sugarcane juice and is Brazil’s most popular distilled alcoholic beverage.

According to Gruber, consumers prefer the flavor and odor of cachaça aged in oak casks, so it sells for a higher price. However, oak is not native to Brazil, and import restrictions often apply.

Native species increasingly used to age cachaça include jatobá (Hymenaea courbaril), jacaranda (Jacaranda mimosifolia), jequitibá (Cariniana estrellensis) and walnut (Ocotea porosa). According to Gruber, the producer may claim that oak has been used.

“Some distillers sell cachaça labeled as aged in jatobá for less than oak-aged spirit, but the opposite is also true: you find cachaça aged in barrels made from native wood but with oak on the label and selling for 70 dollars a bottle,” Gruber said.

To try to protect consumers from this kind of false advertising, the researchers adapted IQ-USP’s e-nose for the analysis of cachaça samples. “The device is able to ‘sniff’ a shot of aged cachaça and identify the wood species used,” Gruber said.

This particular e-nose was developed during the post-doctoral project “Distinguishing hydroalcoholic wood extracts and monitoring their stages of aging using gas sensors, gas chromatography (GC-MS) and multivariate analysis,” supported by a grant from FAPESP.

Plastic identification

The researchers at IQ-USP also developed a device to identify plastics for recycling.

According to Gruber, different types of plastic, such as polyethylene and polypropylene, cannot be mixed when recycled because they contain incompatible resins.

One of the techniques used to identify and classify plastics is infrared spectroscopic analysis of samples dissolved in appropriate solvents.

However, infrared spectroscopy requires a laboratory staffed by professionals qualified to operate relatively sophisticated equipment. The e-nose developed by Gruber’s research group is much simpler than a spectrometer and cheaper to operate. It identifies plastics from the gas emitted in combustion.

The researchers built a small combustion chamber to hold a sample of approximately 300 milligrams for incineration.

The e-nose “sniffs” the fumes emitted during combustion and identifies the type of plastic from the volatile compounds that the plastic releases.

“Polyethylene produces carbon dioxide and water during combustion, whereas a polyamide like nylon, for example, produces nitrogen oxides as well as carbon and water. The e-nose perceives those differences” Gruber said.

Fungal contamination

The researchers also developed an e-nose for early detection of contamination of oranges (after being harvested) by Penicillium digitatum.

This fungus, together with Elsinoe australis and Guignardia citricarpa, causes severe economic losses in countries that are major citrus producers, such as Brazil. The e-nose can detect contamination by the fungus while oranges are in silo storage and before it is visible.

“The device detects contamination as soon as day two and, in a matter of seconds, identifies infection of the oranges by the fungus based on the volatile metabolites that it releases,” Gruber said.

Commercialization

According to Gruber, some of the e-noses developed by his group are protected by patents. The idea is for interested firms to license the technology to produce the devices and sell them.

“We aim to produce low-cost e-noses. Some of the commercially available devices cost as much as 20,000 dollars,” Gruber said. One of the reasons for the high price, he explained, is that these devices have 20-30 sensors and are not designed for specific applications.

“We develop e-noses for specific applications, so we can reduce the number of sensors that they contain and greatly reduce production cost,” he said.

The paper “New composite porphyrin-conductive polymer gas sensors for application in electronic noses” (doi: 10.1016/j.snb.2013.11.022), by Gruber et al., can be read in Sensors and Actuators B: Chemical at www.sciencedirect.com/science/article/pii/S0925400513013646.

The paper “A conductive polymer based electronic nose for early detection of Penicillium digitatum in post-harvest oranges” (doi: 10.1016/j.msec.2013.02.043), also by Gruber et al., can be read in Materials Science and Engineering C at www.sciencedirect.com/science/article/pii/S0928493113001434.